hZzy commited on
Commit
b8ff1de
1 Parent(s): e0d3a9c

End of training

Browse files
Files changed (4) hide show
  1. README.md +16 -9
  2. all_results.json +14 -0
  3. config.json +1 -1
  4. eval_results.json +17 -0
README.md CHANGED
@@ -2,9 +2,16 @@
2
  license: apache-2.0
3
  base_model: hZzy/qwen2.5-0.5b-sft-news-IFT
4
  tags:
 
 
5
  - trl
6
  - expo
7
  - generated_from_trainer
 
 
 
 
 
8
  model-index:
9
  - name: qwen2.5-0.5b-expo-L2EXPO-ES-0.01
10
  results: []
@@ -16,18 +23,18 @@ should probably proofread and complete it, then remove this comment. -->
16
  [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/zhiyuzha-university-of-florida/huggingface/runs/8q1vl3fv)
17
  # qwen2.5-0.5b-expo-L2EXPO-ES-0.01
18
 
19
- This model is a fine-tuned version of [hZzy/qwen2.5-0.5b-sft-news-IFT](https://huggingface.co/hZzy/qwen2.5-0.5b-sft-news-IFT) on an unknown dataset.
20
  It achieves the following results on the evaluation set:
21
- - Loss: 0.3975
22
- - Logps: -123.3866
23
- - Logits: -2.2621
24
- - Objective: 0.3953
25
- - Dpo Loss: 0.6769
26
- - Regularize: 0.3953
27
- - Ranking Simple: 0.5740
28
  - Ranking Idealized: 0.8732
29
  - Ranking Idealized Expo: 0.5321
30
- - Wo Beta: 23.6726
31
 
32
  ## Model description
33
 
 
2
  license: apache-2.0
3
  base_model: hZzy/qwen2.5-0.5b-sft-news-IFT
4
  tags:
5
+ - alignment-handbook
6
+ - ndcg
7
  - trl
8
  - expo
9
  - generated_from_trainer
10
+ - trl
11
+ - expo
12
+ - generated_from_trainer
13
+ datasets:
14
+ - hZzy/train_pairwise
15
  model-index:
16
  - name: qwen2.5-0.5b-expo-L2EXPO-ES-0.01
17
  results: []
 
23
  [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/zhiyuzha-university-of-florida/huggingface/runs/8q1vl3fv)
24
  # qwen2.5-0.5b-expo-L2EXPO-ES-0.01
25
 
26
+ This model is a fine-tuned version of [hZzy/qwen2.5-0.5b-sft-news-IFT](https://huggingface.co/hZzy/qwen2.5-0.5b-sft-news-IFT) on the hZzy/train_pairwise dataset.
27
  It achieves the following results on the evaluation set:
28
+ - Loss: 0.4116
29
+ - Logps: -92.3444
30
+ - Logits: -1.6667
31
+ - Objective: 0.4097
32
+ - Dpo Loss: 0.6906
33
+ - Regularize: 0.4097
34
+ - Ranking Simple: 0.5290
35
  - Ranking Idealized: 0.8732
36
  - Ranking Idealized Expo: 0.5321
37
+ - Wo Beta: 17.6306
38
 
39
  ## Model description
40
 
all_results.json CHANGED
@@ -1,5 +1,19 @@
1
  {
2
  "epoch": 0.8502598016060463,
 
 
 
 
 
 
 
 
 
 
 
 
 
 
3
  "total_flos": 0.0,
4
  "train_loss": 0.3342974317073822,
5
  "train_runtime": 8266.6282,
 
1
  {
2
  "epoch": 0.8502598016060463,
3
+ "eval_dpo_loss": 0.6905742287635803,
4
+ "eval_logits": -1.6666735410690308,
5
+ "eval_logps": -92.34442901611328,
6
+ "eval_loss": 0.41158440709114075,
7
+ "eval_objective": 0.4097130298614502,
8
+ "eval_ranking_idealized": 0.8731883764266968,
9
+ "eval_ranking_idealized_expo": 0.5320910811424255,
10
+ "eval_ranking_simple": 0.5289855003356934,
11
+ "eval_regularize": 0.4097130298614502,
12
+ "eval_runtime": 307.9574,
13
+ "eval_samples": 5790,
14
+ "eval_samples_per_second": 18.801,
15
+ "eval_steps_per_second": 1.568,
16
+ "eval_wo_beta": 17.63062286376953,
17
  "total_flos": 0.0,
18
  "train_loss": 0.3342974317073822,
19
  "train_runtime": 8266.6282,
config.json CHANGED
@@ -23,7 +23,7 @@
23
  "tie_word_embeddings": true,
24
  "torch_dtype": "float32",
25
  "transformers_version": "4.42.0",
26
- "use_cache": false,
27
  "use_mrope": false,
28
  "use_sliding_window": false,
29
  "vocab_size": 151665
 
23
  "tie_word_embeddings": true,
24
  "torch_dtype": "float32",
25
  "transformers_version": "4.42.0",
26
+ "use_cache": true,
27
  "use_mrope": false,
28
  "use_sliding_window": false,
29
  "vocab_size": 151665
eval_results.json ADDED
@@ -0,0 +1,17 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 0.8502598016060463,
3
+ "eval_dpo_loss": 0.6905742287635803,
4
+ "eval_logits": -1.6666735410690308,
5
+ "eval_logps": -92.34442901611328,
6
+ "eval_loss": 0.41158440709114075,
7
+ "eval_objective": 0.4097130298614502,
8
+ "eval_ranking_idealized": 0.8731883764266968,
9
+ "eval_ranking_idealized_expo": 0.5320910811424255,
10
+ "eval_ranking_simple": 0.5289855003356934,
11
+ "eval_regularize": 0.4097130298614502,
12
+ "eval_runtime": 307.9574,
13
+ "eval_samples": 5790,
14
+ "eval_samples_per_second": 18.801,
15
+ "eval_steps_per_second": 1.568,
16
+ "eval_wo_beta": 17.63062286376953
17
+ }